PARP: Prune, Adjust and Re-Prune for Self-Supervised Speech Recognitiont
Link : https://openreview.net/pdf?id=UoVpP8R2Vn 저자/학회 특이사항 Neurips 2021 Spotlight paper 주저자는 MIT CSAIL PARP unstructured mag...
Posts by Category
Paper Review
Sparse is Enough in Scaling Transformers
Link : https://arxiv.org/pdf/2111.12763.pdf 저자/학회 특이사항 Google Research in Europe, Reformer 저자도 있고 (Łukasz Kaiser), Performer 저자도 있고(Afroz Mohiu...
Prune Once for All: Sparse Pre-Trained Language Models
Link : https://arxiv.org/pdf/2111.05754.pdf 저자/학회 특이사항 Intel Labs, Israel ENLSP NeurIPS Workshop 2021 제안하는 방법에 대해서는 의구심이 조금 생기지...
Non-deep Networks
Link : https://arxiv.org/abs/2110.07641
MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer
Link : https://arxiv.org/abs/2110.02178?context=cs.LG
Compressing Neural Networks: Towards Determining the Optimal Layer-wise Decomposition
Link: https://arxiv.org/pdf/2107.11442.pdf
Understanding and Overcoming the Challenges of Efficient Transformer Quantization
Link : https://arxiv.org/abs/2109.12948v1
Block Pruning For Faster Transformers
Link : https://arxiv.org/abs/2109.04838